Immaculate Disinflation

Debbie Johnson & Ed Yardeni

This is an excerpt from Yardeni Research Morning Briefing, July 17, 2024.

During the spring of 2022, Debbie and I predicted that inflation was peaking: “In our scenario, the PCED headline inflation rate peaks during H1-2022 between 6%-7%. Led by consumer durable goods prices, it moderates to 4%-5% during H2-2022. Next year, it falls to 3%-4% as persistently rising rent inflation offsets moderation in other consumer prices.” We wrote that in our April 19, 2022 Morning Briefing . We expected that goods inflation would decline faster than rent inflation. In our September 11, 2023 Morning Briefing , we predicted that inflation would fall to 2%-3% in 2024.

We first wrote about “immaculate disinflation” in the September 6, 2022 Morning Briefing :

“Is immaculate disinflation possible? History shows that inflation rarely falls on its own without a recession. But we don’t think history necessarily has to repeat itself (despite how often it rhymes). … What seems to be different this time (so far) is that the credit system is less vulnerable to a credit crunch than it was in the past. The result is what we now have: a rolling recession hitting different sectors of the economy at different times; we expect it to bring inflation down without precipitating an economy-wide downturn.”

Since the start of the year, we’ve forecast that the PCED inflation rate will fall to 2.0% by the end of this year (Fig. 11 ). So far, so good. June’s CPI report suggests we are still on the right track:

(1) CPI vs PCED. The CPI is released by the Bureau of Labor Statistics (BLS) a couple of weeks before the PCED comes out for each month. The PCED measure is compiled by the Bureau of Economic Analysis (BEA). The components of the CPI are used to calculate the PCED. Most of the component series are identical. A few differ because different methods are used to estimate them by the BLS and by the BEA. Many of the component series have different weights.

The core CPI inflation rate closely tracks the core PCED inflation rate. However, on average since 1960, the former has exceeded the latter by 0.5 percentage points (Fig. 12 ). During May, the core CPI was up 3.4% while the core PCED rose 2.6%. June’s core CPI was released last week showing an increase of 3.3%.

(2) Durable goods. Much of the discrepancy between the two consumer price measures is attributable to durable goods consumer prices, which have increased 1.0 percentage point faster in the CPI than in the PCED since 1960 (Fig. 13 ). That might be mostly attributable to “hedonic” price adjustments in PCED durable goods prices to reflect the fact that today’s durable goods have numerous features that improve on those of older versions.

During May, the durable goods CPI was down 3.8% while the durable goods PCED fell 3.2%. June’s durable goods CPI was released last week showing a decrease of 4.1%.

(3) Medical care services. Another significant divergence between the CPI and PCED core inflation rates is attributable to medical care services (Fig. 14 ). The former has exceeded the latter by 0.7 percentage points since 1960. During May, the CPI and PCED measures of medical care services both rose 3.1% in May. June’s CPI measure of this component was up 3.3%.

There’s a big discrepancy between hospital fees in the CPI and PCED (Fig. 15 ). The former reflects the out-of-pocket costs paid by consumers, while the latter also reflects the costs that are covered by health care insurance and government health care programs.

(4) Health insurance. Another major wild card is health insurance. The CPI measure is convoluted and extremely volatile. The PCED measure is much less so and more sensible.

(5) With and without shelter. Last but not least, shelter inflation is finally moderating too, boding well for overall inflation. As noted above, June’s headline and core CPI fell 0.1% m/m (3.0% y/y) and rose 0.1% m/m (3.3% y/y). Excluding shelter inflation, both indexes are down to only 1.8% y/y (Fig. 16 )!

Rent inflation as measured in the CPI is falling, catching up to lower market rent inflation. Shelter (which accounts for one-third of the CPI) rose just 0.2% m/m (5.2% y/y) in June, its slowest monthly pace since January 2021. The three-month percentage change in shelter’s two major components—rent of primary residence and owners’ equivalent rent—suggest that their y/y readings will continue to fall (Fig. 17 ).

Try our research service . See our Predicting the Markets book series on Dr. Ed's Amazon Author Page . Please see our hedge clause .